AbstractNetworks are widely used representation of complex systems. Using network representations, both topologies and dynamics of gene regulation systems in different species and cell types were studied in this thesis. The topological structure of linear regulatory chains (LRCs) within gene regulatory networks (GRNs) were explored. Network analyses were conducted using a combination of mathematical modelling, statistical simulation and large-scale data analysis. Our results indicated LRCs to be regulatory attenuators in GRNS, and were found to be absent from or rare in GRNs of E. coli K12, Mycobateria tuberculosis, yeasts and human non-cancer cells. However, they are enormous in human cancer cells. In addition, LRCs always interact with chaotic motifs in GRNS to regulate gene expressions in all the species and cell types examined. Effects of gene dosages on transcriptional dynamics of three GRN motifs were studied using both deterministic and stochastic models and the results suggested that gene dosages affect strongly behaviours of transcriptional motifs in GRNs and potentially promote heterogeneity in cell populations. When looking at transcriptional behaviours of the human blood stream parasite Trypanosoma brucei using mathematical models, the population dynamics of the parasites were found to be associated with the lengths of a family of surface antigen genes. This dynamic behaviour was interpreted as a ‘feint attack’ diversion tactic utilised during infection by these persistent parasites, allowing the infection to out-maneuver the host immune system. In addition to investigating existing biological networks, a matrix factorization method was adapted to produce enhancerpromoter networks for human reference epigenomes. This was achieved by integrating data from different sources and of different qualities and predicts regulatory networks of chromatin interaction edges linking more than 20,000 promoters and 1.8 million enhancers across 127 human cell and tissue types. Despite the diverse biological topics covered in this thesis, all the projects share the fundamental theme of using networks to present and simple modelling to understand complex biological systems.
|Date of Award||2017|
|Add any sponsors of the thesis research||Wellcome Trust, Vest Scholarship & Massachusetts Institute of Technology|
|Supervisor||Timothy Newman (Supervisor)|
Modelling biological networks : topology, dynamics and generation
Liu, D. (Author). 2017
Student thesis: Doctoral Thesis › Doctor of Philosophy